posted by Den .
Simply reporting measures of central tendency or measures of variability will not tell the whole story. Using the following information, what else does a psychologist need to know or think about when interpreting this information?
A school psychologist decided to separate some classes by gender to see if learning improved. She looked at student scores on the final exam and obtained the following information: Students in boy-girl classrooms obtained an average of 71.4 on their final exams with a standard deviation of 10.8 whereas students in single-gendered classrooms obtained an average of 75.9 on their final exams with a standard deviation of 8.2. She concludes that the single-gendered classrooms lead to better learning.
I would be curious if one single-gendered classes did better than the other too. I would want to separate into all-boy and all-girl data. The difference between combined and separated data might be due solely to one gender or the other.